(alpha,n) reactions

Report
Report on
The Study of (α,n) Neutron Yield
and Energy Spectrum
Dongming Mei for the AARM
collaboration
1
Motivation
• (α,n) neutrons produced in the materials, which will be
used to build the detector components for low background
experiments, are important backgrounds
• Both neutron yield and energy are important
• Cross sections in particular resonances should be validated
• The calculation done for alphas from 238U and 232Th are not
sufficient
–
228Th
and 226Ra needs to be taken into account
• The yield and energy spectrum in different materials are
only calculated by a few groups
• The validation of the calculations are needed
– Start with independent comparisons
– Measurements and benchmarks are also on the way
2
Calculations by the SNO collaborators
• R. Heaton et al., Nucl. Geophys. V 4, 499
(1990).
• R. Heaton et al., Nucl. Instrum. Methods Phys.
Res. A 276, 529 (1989).
3
USD Calculations
• D.M.Mei, C.Zhang, A.Hime, NIMA606, 651-660
(2009)
• Http://neutronyield.usd.edu
4
Calculations using SOURCES
• http://www.ornl.gov/info/reports/1992/3445
603684010.pdf
• Radial Prot Dosimetry, 2005: 155 (1-4) 117-21
5
Comparison (USD vs
SOURCES)Implemented By European
Scientists
• Marco Selvi (one of several)
The overall agreement in the neutron yield is
anyhow quite good ,I found everything within
a factor of 2, which is not so bad I would say.
6
Comparison Accomplished by US
Scientists
• K.Palladino (MIT), H.Qiu (SMU), S.Scorza
(SMU)
7
Modus Operandi
• TENDL vs SOURCES4 lib cross sections
• TENDL 2011 and 2012 have been considered as
the USD website inputs. TENDL is a nuclear data
library (validated) which provides the output of the
TALYS nuclear model code system
• SOURCES4 cross section input libraries come
from EMPIRE calculations and for some isotopes a
combination of measurements and EMPIRE
calculations
• USD website vs SOURCES4 calculations:
compare the radiogenic neutron spectra coming
from both codes and some simulation quick
checks for Cu.
8
X section comparison
✔Good agreement in the (alpha,n) ROI (0-10MeV) for most of
the isotope considered
– details http://www.physics.smu.edu/cooley/aarm/webpage.html
✖C13, O17, … : SOURCES4 inputs match TENDL 2011 at low
energy and then match TENDL 2012 at high energy
Cu65 -> ok!
C13
SOURCES4 input – TENDL 2011 –TENDL 2012
9
X section input libraries
For many isotopes SOURCES4 cross section input libraries
consist in a combination of measurements and EMPIRE
calculations -> we believe SOURCES4 cross section
libraries the right choice.
- EMPIRE is the code recommended by IAEA.
- Neither EMPIRE nor TALYS can calculate properly
resonance behavior which has been experimentally
observed (if we trust the data)
10
Copper check
- Inputs
• SOURCES4 calculation considers
– 63Cu = 70%, 65Cu = 30%
– 1ppb 232Th in Cu (100% 232Th)
– 1ppb U in Cu (99.28% 238U + 0.72% 235U)
• USD website considers
– Nat Cu
– 1ppb 232Th in Cu (100% 232Th)
– 1ppb U in Cu (100% 238U)
11
Copper Check
- Neutron Spectrum Comparison
12
Spectra Integration (n/s/cm3)
SOURCES4
Th: 9.49 E-12 n/s/cm3
U: 2.90 E-12 n/s/cm3
USD website
Th: 1.11 E-11 n/s/cm3
U: 3.46 E-12 n/s/cm3
SOURCES4/USD Discrepancies
Thorium ~15%
Uranium ~13%
13
Simulation check
We have performed some quick simulations: propagate both
USD and SOURCES4 radiogenic neutron spectra in the same
experimental geometry to check the background neutron rate
• 1 Million neutrons from U and Th decay chains each in a
simple geometry: 1 Cu can (21760.6 cm3) around 100kg of
germanium detector.
• The same Cu can has been contaminated with both USD and
SOURCES spectra
• Cu contamination level:
Th: 0.02mBq/kg
U: 0.1 mBq/kg
• Neutron rate from (alpha,n) reactions and has been reported
14
SOURCES4/USD ~20% discrepancy
in the rate found in Ge detectors
USD rate > SOURCES4 rate
15
Borosilicate glass check
• Glass composition from Hamamatsu, more elements and higher
neutron yields than just silicon, oxygen and boron due to inclusion of
4% each by mass fraction inclusion of sodium, aluminum and barium
16
Borosilicate Spectra Integration
(n/s/cm3)
SOURCES4
Th: 1.27 E-10 n/s/cm3
U: 3.63 E-10 n/s/cm3
USD website
Th: 6.98 E-11 n/s/cm3
U: 2.45 E-10 n/s/cm3
SOURCES4/USD Discrepancies
Thorium ~82%
Uranium ~48%
17
Borosilicate Simulations
• Simulations done within RAT, utilizing Geant4.9.5, and a cylindrical
45T liquid argon single phase detector surrounded by borosilicate
glass mimicking PMTs as well as stainless steel and a water veto
• Simulation done with 1.65 times more thorium than uranium
matching assayed values of glass (and old simulations)
# Simulated
Events 12-25 & >65 cm
keVee
from wall
& PSD cut MC single
scatters in ROI
USD
5000000
11651
211
9
2
Sources
5000000
12133
217
10
2
Borosilicate Conclusion: Shape differences are not a large effect, but at
this assay value Sources would have 1.8x more neutrons produced than
the USD code
18
Next steps
• Check SOURCES calculations having Talys
cross section as inputs
• Check USD calculations having Empire
inputs (is it possible?)
• Cross check EMPIRE and TALYS cross
sections with other calculations
• Benchmarking against the experimental nuclear
data.
(SOURCES has already provided some comparison
studies in the users guide but not for U/Th decay
chains)
19
Conversion factor #->n/kg/y
Here below the formula used for calculating the normalization factor
needed to convert the number of neutron found in the IZip from
simulation into a counting rate (n/kg/year).
20
Summary of the Comparisons
• Neutron yield agreement is within a factor of
2 for all materials compared so far
• The agreement in energy spectra from various
materials is not good
– Understand the cross sections including
resonances
– The calculated kinetic energy of out-going
neutrons with respect to different excited states of
the final nucleus
21
Neutron Screening Facility (LZ-veto)
Gd-LS detector (Courtesy to the LZ collaboration))
22
LZ-Veto as a Neutron Screening
The goals of this device are:
• To further characterize the neutron and gamma ray environment adjacent
to the LUX liquid xenon, so as to better constrain possible background
contributions to any apparent nuclear recoil signal from WIMPs.
• To screen a variety of components for neutron and gamma activity. Of
particular interest are components containing fluorine like
polytetrafluoroethylene (PTFE) used in noble-liquid dark matter detectors,
including the LUX liquid xenon detector. Fluorine is particularly
susceptible to emission of neutrons caused by impinging alpha particles,
from radioactive contamination both on its surface and intrinsically. Such
neutrons can potentially contribute to the background for a WIMP signal.
In addition, this device can screen some materials more accurately for
gamma-ray emission than standard germanium screening devices.
• To develop and establish safe and effective designs and procedures for
deployment of Gd-LS deep underground.
• To conduct searches for rare processes involving bursts of neutrons, for
example, the spontaneous fission of 232Th, or searches for super-heavy
elements.
23
Future Work
• Continue to work internationally as a collaboration
• Monte Carlos Validation
–
–
–
–
Energy spectra in different materials
Specifically focus on resonances in cross section
Include 228Th and 226Ra
Materials such as PTFE
• Measurements
– USD neutron detector for measuring the (α,n) neutrons
from rock
– Other measurements from various experiments
– LZ-veto for measuring (α,n) the neutrons from various
materials
24

similar documents